Instantaneous SmI2-H2O-mediated reduction of dialkyl ketones induced by amines in THF

Article abstract of DOI:10.1016/S0040-4039(02)01673-8

Reduction of various ketones to their corresponding alcohols is shown to be instantaneous, i.e. completed in less than 10 s, by samarium diiodide (2.5 equiv.) in the presence of water (6.25 equiv.) and an amine (5 equiv.) in THF. The rates of reduction of ketones in this mixture exceed by far the rates determined by an amine or water alone. Rate enhancement is at least 100 000 compared to the reduction without a proton source, or at least 100 times faster than the rate of the widely used HMPA/alcohol accelerated reductions. This new method is therefore suggested to be an excellent replacement of the toxic HMPA/alcohol method.

Full text of DOI:10.1016/S0040-4039(02)01673-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 7197–7200
Instantaneous SmI2–H2O-mediated reduction of dialkyl ketones
induced by amines in THF
Anders Dahle´n and Go¨ran Hilmersson*
Organic Chemistry, Department of Chemistry, Go¨teborg University, SE-412 96 Go¨teborg, Sweden
Received 27 May 2002; accepted 9 August 2002
Abstract—Reduction of various ketones to their corresponding alcohols is shown to be instantaneous, i.e. completed in less than
10 s, by samarium diiodide (2.5 equiv.) in the presence of water (6.25 equiv.) and an amine (5 equiv.) in THF. The rates of
reduction of ketones in this mixture exceed by far the rates determined by an amine or water alone. Rate enhancement is at least
100 000 compared to the reduction without a proton source, or at least 100 times faster than the rate of the widely used
HMPA/alcohol accelerated reductions. This new method is therefore suggested to be an excellent replacement of the toxic
HMPA/alcohol method. © 2002 Elsevier Science Ltd. All rights reserved.
The chemistry of samarium diiodide (SmI2) in THF has
developed rapidly during recent years to become a
reagent of particular importance as a mild reducing
reagent.1 The popularity of SmI2 as a selective reducing
reagent has been largely dependent on the increased
reactivity found upon addition of HMPA.2 It has been
shown that HMPA increases the oxidation potential of
SmI2 from −1.33 to −2.05 V.3 The accelerations of
ketone reductions are reported to result in reaction
times of about 10–30 min instead of days.1a However, it
would be desirable to avoid the use of the carcinogenic
compound HMPA.
The initial step in SmI2-mediated reductions of ketones
is believed to be the transfer of one electron yielding a
ketyl radical. The fast equilibrium between ketone and
ketyl radical is believed to be lying on the left side,
based on reported mechanistic evidence (Fig. 1).7,8 Our
idea is to drive the equilibrium forward by means of
precipitation of by-products, e.g. the ions Sm3+ and I−.
Herein we wish to report the use of amine/water mix-
tures in the SmI2-mediated reduction of ketones into
alcohols.
Earlier studies of SmI2-mediated reductions have shown
that an amine3 or water4 alone can be used to enhance
the rates. However, there are no reports on the combi-
nation of these two additives, which bring about a
remarkable increase in rate. The rate is far beyond the
expected, i.e. the sum of the rates of water and amine,
respectively.
Proton sources such as methanol or t-butanol are
known to facilitate SmI2-mediated reductions.5
Recently we reported on the SmI2-mediated reduction
of 3-heptanone accelerated by diols with coordinating
ethereal groups.6 We showed that the rate of reduction
of the ketone was accelerated some 255 times upon the
addition of diglycol, a diol with three oxygens, capable
of coordinating to samarium. The proton transfer was
found to be rate determining with a primary kinetic
isotope effect (kH/kD) of 2.
Figure 1. General mechanism for conversion of a ketone into
an alcohol in SmI2-mediated reductions.7,8
During our kinetic studies using various proton donors
we were concerned with the difficulty of completely
removing all traces of water from the additives, there-
* Corresponding author. Tel.: 46 31 772 2904; fax: 46 31 772 3840;
e-mail: hilmers@organic.gu.se
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01673-8

7198
A. Dahle´n, G. Hilmersson / Tetrahedron Letters 43 (2002) 7197–7200
The effect of alcohol addition in place of water in
combination with the amine did not give these extraor-
dinary rate enhancements (entry 4). We therefore con-
clude that the rate enhancement for the SmI2-mediated
reduction must be a cooperativity effect of both amine
and water (Scheme 1).
fore several experiments with deliberate addition of
water were performed. Water alone gave only a small
rate enhancement (less than a factor of 10) compared to
the SmI2-mediated reduction in THF without any pro-
ton source. Tertiary amines appear to retard the reduc-
tion. To our surprise we observed rate enhancements
considerably larger for a triethylamine/H2O mixture
than expected, i.e. the sum of the water and the amine
accelerated SmI2-mediated reductions, respectively. The
reduction of 3-heptanone was complete in just a few
seconds.† To investigate further this interesting discov-
ery various amines were used as additives together with
water. In organoalkali chemistry the diamine
N,N,N%,N%-tetramethylethylenediamine (TMEDA) is
often used to enhance the rate of metallation reactions.9
Since samarium(II) is regarded as a rather hard Lewis
acid it was a natural choice to investigate the effect of
TMEDA on the SmI2–H2O-mediated reduction of 3-
heptanone. We observed that TMEDA/H2O mixtures
also gave a very large rate enhancement. In our previ-
ous report on diol accelerated reduction of 3-heptanone
we could evaluate the different alcohols by comparing
the initial rates of reduction.6 However, with the amine/
water mixtures we observed that the reductions were
instantaneous and the initial rates could not be deter-
mined. Instead a qualitative measurement of the reac-
tion rate was obtained from the analysis of the reaction
mixture after 1 min. By analogy with our earlier results
of the alcohols and diols we also investigated the
triamine, N,N,N%,N¦,N¦-pentamethyldiethylenetriamine
(PMDTA) in the SmI2–H2O-mediated reduction. Again
100% of the alcohol was obtained after less than 1 min
of reaction time (Table 1). To ensure complete reduc-
tion excess water (14 equiv.) was used in combination
with the amine.
Scheme 1.
The extremely fast reactions (entries 6, 8 and 9 in Table
1) were also investigated after 10 s, which is the shortest
time to allow mixing of the reagents in the Schlenk
tube. Only the alcohol was detected even after this
short reaction time for entries 8 and 9 indicating that
the mixing of H2O, amine, SmI2 and ketone is rate
limiting. A rough estimate of the rate based on this
indicates at least 10% yield in less than 1 s, provided a
linear dependence of yield versus time over the whole
range from 0 to 100%. The initial rate is larger as the
rate must fall off towards completion. However, this
estimation corresponds to a rate enhancement of more
than 100 000 in comparison with the SmI2-mediated
reaction in the absence of a proton donor. SmI2-medi-
ated reduction of ketones induced by HMPA/alcohol is
reported to give 95% yield in 10–30 min in comparison,
i.e. at least 700–1000 times faster than that without
additives.1,2 This means that our amine/H2O method is
more than 100 times faster than the HMPA/alcohol
method.
We observed that a minimum of 2.5 equiv. of commer-
cial purchased 0.1 M SmI2 in THF was needed in order
Table 1. Different proton donors and additives together with the respective yield (determined by GC) of 3-heptanol in the
SmI2 (7 equiv.)-mediated reduction of 3-heptanone (1 equiv.) in THF after 1 min of reaction time
Entry
Proton source (equiv.)
Ligand (equiv.)
Yield (%)
Relative ratea
1
2
3
4
5
6
7
8
9
–
–
–
–
0.0076
0.0186
1.76
1
2.6
255
100
0.9
H2O (7)
Diglycol (7)
MeOH (14)
–
H2O (14)
–
Et3N (7)
Et3N (7)
Et3N (7)
TMEDA (7)
TMEDA (7)
PMDTA (7)
0.7
0.006b
\99
\100 000c
0.4
0.003d
\99
H2O (14)
H2O (14)
\100 000c
\100 000c
\99
a Based on initial rates.
b Estimated value, based on the yield (1.5%) after 3 h.
c Initial rate could not be determined, due to 100% yield in 10 s.
d Estimated value, based on the yield (0.6%) after 3 h.
† In a standard procedure, SmI2 in THF (5.0 ml, 0.1 M, Aldrich) was added to a dry Schlenk tube, fitted with a septum and containing a magnetic
stirrer bar, inside a glove box with nitrogen atmosphere. The proton donor, i.e. H2O, and the amine was added under stirring. To this mixture
was then added the ketone, e.g. 3-heptanone (10 ml, 0.14 equiv.), at 20.0°C. Small portions of the mixture (100 ml) was removed via a syringe
and quenched with I2 in n-hexane (0.1 M, 0.1 ml) including 1-hexanol (0.016 M) as internal standard. To the quenched solution was added
diethyl ether (1 ml) and HCl (0.12 M, 0.1 ml) to dissolve the inorganic salts and finally Na2S2O3 to remove excess iodine. The organic layer was
transferred to a vial and analyzed by GC. All products were analyzed by 1H NMR, GC/MS and also compared with references on GC.

A. Dahle´n, G. Hilmersson / Tetrahedron Letters 43 (2002) 7197–7200
7199
to obtain quantitative yield of 3-heptanol. Adding the
amine and H2O in portions proved that 2 equiv. of
Et3N and 3 equiv. of H2O for each molecule of reacted
SmI2 must be used in order to obtain 100% yield in the
reduction (Table 2). The reaction appears to be instan-
taneous based on the colour change from blue to white,
reflecting the oxidation of Sm(II) to Sm(III). A clear,
colourless solution and a white precipitate was obtained
in less than 10 s. It was also demonstrated that these
ratios of SmI2, H2O and amine were optimal using the
bi- and tridentate amines. However, as the diamines
have two amino functions only 1 equiv. of diamine for
each SmI2 was required. For the triamine only 0.7
equiv. for each SmI2 was necessary.
Based on our results we propose that samarium diio-
dide is coordinated by two nitrogens of one molecule of
the diamine or two molecules of monoamine. Dimeric
species of the type [SmI2(R3N)2(OH)2]2 or [SmI2-
(TMEDA)(OH)2]2 may be key species in solution. It
appears that the presence of H2O and amine promote
rapid precipitation of insoluble salts, for instance
Sm(OH)3 and R3N·HI.‡ Our results also confirmed that
it is not sufficient to precipitate only one of the two
proposed insoluble salts of Sm3+ and I− formed to
obtain the maximum rate and yield. This was also
verified in Table 2.
Table 2. Evaluation of the required amount of water and
amine together with the respective yields (determined by
GC) of 3-heptanol in the SmI2 (2.5 equiv.)-mediated
reduction of 3-heptanone (1 equiv.) in THFa
Entry
Proton source (equiv.)
Amine (equiv.) Yield (%)
1
2
3
4
5
6
H2O (5.0)
H2O (2.5)
H2O (5.0)
H2O (6.25)
H2O (6.25)
H2O (6.25)
Et3N (2.5)
Et3N (5.0)
Et3N (5.0)
Et3N (5.0)
TMEDA (2.5) \99
PMDTA (1.7) \99
42
41
86
\99
Figure 2. Suggested mechanism for the reduction of ketones
mediated by SmI2, amine and H2O.
a Entries 1–3 stopped at this yield and did not go to completion.
A number of experiments were performed on other
dialkyl ketones (a–h in Scheme 2) to verify that this
method is general. As a general procedure SmI2 (2.5
equiv.), TMEDA (2.5 equiv.) and ketone (1 equiv.)
were mixed followed by the addition of H2O (6.25
equiv.). The reduction of all of these substrates was
clean and instantaneous, i.e. completed in less than
10 s.
A potential mechanism for the SmI2–H2O-mediated
reduction of ketones catalyzed by a bidentate amine is
also presented above (Fig. 2). The exceptionally fast
reaction indicates that it proceeds within one complex,
most likely a dimer of SmI2 since two electrons are
required for the reduction of a ketone. The balanced
equation below describes the reduction.
D
ꢁꢁꢁꢁ
R CꢀO+2SmI +6H O+4R N
R2CHOH+
C
2
2
2
3
2Sm(OH)3+4R3N·HI
We have not examined the composition of these samar-
ium and amine salts, which means that other salts may
also be present. However, only trace amounts of amine
could be found in the THF solution indicating that the
amine is existing in the precipitate. Addition of NaOH
(1 M) followed by extraction into diethyl ether (DEE)
showed that the amine could be isolated from the salts,
leaving a white precipitate of water and DEE insoluble
samarium-salts. The work-up of the SmI2-mediated
reductions of ketones can thus be simplified to filtration
of the salts and evaporation of THF.
The large rate enhancement found for the amine cataly-
sis of the SmI2–H2O-mediated reduction of ketones is
proposed to be a result of one or several factors:
‡ Only trace amounts of the amine were detected in the supernatant
after the SmI2-mediated H2O- and amine-catalyzed reactions were
complete.
Scheme 2. Various dialkyl ketones examined in the SmI2-
mediated reduction induced by an amine and H2O.

7200
A. Dahle´n, G. Hilmersson / Tetrahedron Letters 43 (2002) 7197–7200
(1) SmI2 is oxidized and in the presence of water and
amine there is a rapid formation of Sm(OH)3 and a
quaternary ammonium salt R3N·HI. Rapid precipitation
of these salts drives the equilibrium forward. Only trace
amounts of the amine were detected in the supernatant
after the SmI2-mediated H2O and amine-catalyzed reac-
tions were completed.
References
1. (a) Molander, G. A. In Organic Reactions; Reductions
with Samarium(II) Iodide. Paquette, L. A., et al., Eds.;
John Wiley & Sons: New York, 1994; Vol. 46, pp. 211–
367; (b) Krief, A.; Laval, A.-M. Chem. Rev. 1999, 99,
747–777; (c) Molander, G. A.; Harris, C. R. Chem. Rev.
1996, 96, 307–338; (d) Molander, G. A. Chem. Rev.
1992, 92, 29–68; (e) Nomura, R.; Endo, T. Chem. Eur.
J. 1998, 1605–1610; (f) Kim, S. M.; Byun, I. S.; Kim, Y.
H. Angew Chem. 2000, 112, 744–747; (g) Mikami, K.;
Yamaoka, M.; Nakamura, Y.; Takeuchi, S.; Ohira, A.;
Ohgo, Y. Tetrahedron Lett. 1996, 37, 2805–2808; (h)
Takeuchi, S.; Nakamura, Y.; Ohgo, Y.; Curran, D. P.
Tetrahedron Lett. 1998, 39, 8691–8694; (i) Lodberg, H.;
Christensen, T. B.; Enemaerke, R. J.; Daasbjerg, K.;
Skrydstrup, T. Eur. J. Org. Chem. 1999, 565–572; (j)
Steel, P. S. J. Chem. Soc., Perkin Trans. 1 2001, 2727–
2751.
2. (a) Inanaga, J.; Ishikawa, M.; Yamaguchi, M. Chem.
Lett. 1987, 1485; (b) Fuh, M.-R. S.; Lin, T.-Y.; Chang,
S.-C. Talanta 1998, 46, 861–866; (c) Lin, T.-Y.; Fuh,
M.-R.; Chang, S.-C. J. Chin. Chem. Soc. 1999, 46, 581–
584.
3. Shabangi, M.; Sealy, J. M.; Fuchs, J. R.; Flowers, R.
A., II Tetrahedron Lett. 1998, 39, 4429–4432.
4. Girard, P.; Namy, J. L.; Kagan, H. B. J. Am. Chem.
Soc. 1980, 102, 2693–2698.
(2) The increased reducing power of amine coordinated
SmI2, in analogy with additives such as HMPA, DBU or
TMU.3
In conclusion, we have found that amines are effective
additives for the SmI2–H2O-mediated reduction of
ketones. Mono-, bi- and tridentate amines, such as Et3N,
TMEDA and PMDTA, respectively, accelerate the
SmI2–H2O-mediated reduction of ketones; the rates can
be increased more than 100 000 times. The reaction
appears to be instantaneous, i.e. the mixing of the
reagents is rate limiting. This impressive effect now sets
the stage for a new era in SmI2-mediated reductions. The
mixture of water and amine is clearly superior to the
widely used HMPA/alcohol mixtures. The use of amines
in the SmI2–H2O-mediated reduction of ketones is
advantageous because of the impressive rate of reduc-
tion. In addition, the work-up is extraordinarily simple
as all reagents and SmI2 by-products precipitate, leaving
a clean, colourless THF solution with the product
alcohol as the only compound in solution, based on GC
1
and H NMR analysis. These results also indicate that
5. (a) Kagan, H. B.; Namy, J. L.; Girard, P. Tetrahedron
1981, 175–180; (b) Hasegawa, E.; Curran, D. P. J. Org.
Chem. 1993, 58, 5008–5010; (c) Talukdar, S.; Fang, J.-
M. J. Org. Chem. 2001, 66, 330–333.
6. Dahle´n, A.; Hilmersson, G. Tetrahedron Lett. 2001, 42,
5565–5569.
chiral diamines may be promising additives for asymmet-
ric reduction reactions using SmI2 since diamines catalyze
the SmI2–H2O-mediated reduction of ketones. Use of
different chiral amines in asymmetric reduction reactions
mediated by SmI2–H2O are currently under investigation
in our laboratory. Further studies including other func-
tional groups and additives are also under investigation
along with mechanistic studies on the catalytic reaction.
7. Curran, D. P.; Fevig, T. L.; Jasperse, C. P.; Totleben,
M. J. Synlett 1992, 12, 943–961.
8. Prasad, E.; Flowers, R. A., II J. Am. Chem. Soc. 2002,
124, 6357–6361.
9. (a) Collum, D. B. Acc. Chem. Res. 1992, 25, 448–454;
(b) Rutherford, J. L.; Hoffmann, D.; Collum, D. B. J.
Am. Chem. Soc. 2002, 124, 264–271.
Acknowledgements
The Swedish Research Council supported this work
financially.